DSpace Kochi University of Technology
日本語

Kochi University of Technology Academic Resource Repository >
A.学術情報資料別 >
SSMSパブリケーション >
Vol.09 >

Files in This Item:

File Description SizeFormat
sms13-3681.pdf261.48 kBAdobe PDFView/Open
Title: Real-Time Water Resource Policies Supporting System for Decision-Making Based On GPU Application
Authors: BONGOCHGETSAKUL, Nattakorn
SUTTINON, Pongsak
UEMOTO, Kotomi
NASU, Seigo
Issue Date: Dec-2014
Publisher: Society for Social Management Systems
Journal Title: Society for Social Management Systems Internet Journal
Abstract: Climate change is widely recognized to have an adverse effect on both short and long-term, especially under uncertainty of water resource availability. Water is not only affecting drinking and food availability, but also services and industrial manufacturing departments those are significant to our everyday-life. To retain water resource service under such an uncertainty and economics recession circumstance, governor has to determine an appropriate policy from a single or a combination of multiple demand-reduction and supply-increment options, which is not an easy task for the governor. This paper describes designing of a system to answer the best water resource policy under a given constraint conditions, e.g. cost, ease to implement, social and economic scenario, etc. A final solution is determined from considering a vast of number of scenarios from all possible options combinations and uncertainty from climate change prediction, and give the best scenario back to governor in real-time. In each scenario, water resource equilibrium from predicted demand and supply in future is calculated, based on a given (generated) policy, economic and social effects on water shortage are determined. A modern GPU computing and cloud computing technology are integrated in order to serve the most convenient experience to the system users. Adopting GTX-Titan GPU, the system could serve about 1,600,000 scenarios per minute, which is about several ten thousand times faster than the previous system framework. This enable governors using the system to make a decision on the fly.
Type: Conference Paper
URI: http://hdl.handle.net/10173/1236
Appears in Collections:Vol.09

Please use this identifier to cite or link to this item: http://hdl.handle.net/10173/1236

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Kochi University of Technology Library - Feedback